/ About / Archive / Other media
A few points about that study on so-called "call-out culture" - 2/20/18
Conor Friedersdorf, writing for The Atlantic, has called into question what he describes as "the excesses of call-out culture" - particularly on social media - and asks, "what’s your theory of how that could advance social justice?"

As usual, this line of skepticism has prompted a broader discussion on all kinds of distinct questions (is call-out culture good? is it effective? does it ever warrant criticism?) - but here, I just want to touch on one narrow point. In response to the piece, Angus Johnston has pointed readers to an old study that has long circulated among call-out culture's partisans: Condemning and Condoning Racism: A Social Context Approach to Interracial Settings, by Blanchard, This study, Johnston says, establishes that "condemnation of antisocial behavior—and racially offensive speech specifically—is actually a tactic with a proven record of effectiveness".

While that may be true, this is a pretty ambitious reading of what that paper actually says. A few key passages, which its readers routinely neglect:

1. Online is not IRL. Blanchard notes,
Several features of our experimental paradigm may have contributed to the large social influence effects. Two of the elements of social impact theory (LatanĂ©, 1981), strength and immediacy, were set at high levels by our procedures. (996) 
By strength, I mean the...importance, or intensity of a given source to the target - usually this would be determined by such things as the source's status...prior relationship with, or future power over, the target. By immediacy, I mean closeness in space or time and absence of intervening barriers of filters. (344)
Obviously, both of these elements of social influence are experienced quite differently online than they are in Blanchard's experimental setting. Online, the socioeconomic status of the call-outer is often ambiguous or indeterminate; the prior relationship is typically non-existent, as is any "future power" over the target. Similarly, online call-outs are not experienced with any "closeness in space" or in the "absence of intervening barriers or filters." In other words, the very factors that Blanchard says are relevant - strength and immediacy - are both so diminished online that we can't just take for granted the relevance of this study.

All of this predicts quite directly the familiar dismissal of call-outs as coming from "trolls," "bots," "stans," "alts," and so on - critics who do not have status, and who therefore do not have influence. It is also quite consistent with the well-attested Online Disinhibition Effect, which emerges from all of the same factors: anonymity, lack of social proximity, and so on. There is, in short, no reason to assume that social media facilitates influence in the same way that real-world interactions do.

2. Influence declines among anti-racists. This is an effect that Blanchard noticed in a previous study, and that he set out to account for here:
Overhearing another person condemn racism yielded notably robust social influence on the two campuses where those who were exposed to influence were not already uniformly, strongly antiracist. (995, emphasis added.)
This could express what Blanchard calls a "ceiling limitation" (994)- the obvious fact that people who are antiracist (as measured by certain metrics) simply don't have much room to become less racist (as measured by those same metrics).

However - it could also mean that call-outs yield diminishing returns among people who are already "antiracist" in some general sense. I think it's easy to understand why: if you already think of yourself as an antiracist, and are already committed to various points of popular antiracist orthodoxy, you are likely to interpret a call-out as some kind of trivial "intellectual" critique rather than as a compelling attack on values. And as Blanchard notes, "judgmental issues (involving ethical, valued, or proper positions) are more likely to be susceptible to normative influence processes than intellective issues" (996).

If this reading holds, then call-outs may be the least effective where they often seem to emerge the most: in the context of "self-criticism", directed at subjects who see themselves as sharing the same values and same general political beliefs.

3. Influence may not be persistent. Blanchard:
...since we included no means of evaluating the durability of reactions to racism, the longevity of peer influences of the sort we investigated remains unknown. (996)
This point should be of particular concern to activists who are more interested in the long-term fight against racism than against the immediate gratification of fleeting shame or an insincere apology. As I noted previously, other studies have suggested that the "long-term effect" of shaming strategies can entail "a huge impact on one's identity...[it] has a strong impingement on emotional development" (Tanaka) which can precipitate a counterproductive, reactionary response to call-outs.

These points, of course, are just caveats. It may be the case that online call-outs are effective even when we do account for these considerations. And even if they aren't tactically effective, of course, one can still argue for them as expressions of solidarity, as speaking truth-to-power, and so on. But however the evidence shakes out, the case for online call-outs is not well-served by appeals to a study that does not ultimately make that case.